A Feature-Enhanced Deep Learning Network for GPR Hyperbolic Target Detection in B-Scan Profiles

Xin Wang, Qinghe Zhang, Han Liu · 2026

Read A Feature-Enhanced Deep Learning Network for GPR Hyperbolic Target Detection in B-Scan Profiles by Xin Wang, Qinghe Zhang, Han Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13611288.txt.

Description

Xin Wang, Qinghe Zhang, and Han Liu report a 2026 study of a faster way to find buried objects in radar scans. Ground radar shows buried targets as curved marks in its scan images. Those marks are small, faint, and lost in clutter, so they are hard to spot. The team adapted a fast image detection model for the task. They added parts shaped to catch curved edges and parts that cut wasted computing.

On a data set the team built for the task, the model scored high on both exactness and coverage. It also beat the base model while using fewer settings and less computing. Its speed was well above what real-time work needs. The trade-off the authors aimed for, high scores at low cost, was met in their tests. Such a tool could help crews scan roads, pipes, and sites with less hand review.

Catalog record

CREATORXin Wang, Qinghe Zhang, Han Liu
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13611288

Browse the full library · About the Tower of Babel